# Artsdata MCP Server > Machine-readable guide for LLM and agent tooling. ## Programmatic / agent access - MCP server endpoint: https://mcp.artsdata.ca/mcp - Transport: Streamable HTTP (JSON-RPC 2.0 over POST) - Session model: stateless - Server instructions: Access Artsdata's public APIs through a secure, lightweight, LLM-optimized layer designed for efficient querying by autonomous agents and users alike. SPARQL endpoint: https://query.artsdata.ca/query. To answer a question with SPARQL, call get_schema to learn the data model, write the query and run it with sparql_query, then use get_entity (details of a URI) or search_entities (URI from a name) as needed. ## MCP tools ### search_entities Tool to search for entities in the Artsdata knowledge graph by name, with optional filters on entity type and on the controlled vocabulary (skos:ConceptScheme) a concept belongs to. Input parameters - query (string, required): Non-empty mandatory search query. - types (array, optional): Optional list of type filters; can be empty. - in_scheme (array, optional): Optional list of Artsdata controlled vocabulary (skos:ConceptScheme) URIs (e.g., ['http://kg.artsdata.ca/resource/ArtsdataEventTypes']). Paired with types: ['http://www.w3.org/2004/02/skos/core#Concept']. Use it to find a concept URI by its label (e.g., the event type for 'exhibition') for use in search_events under has_event_type_concept. Can be left empty. - language (string, optional, default en, enum en|fr): Language code, restricted to 'en' or 'fr'. - limit (integer, optional, default 25, min 1, max 50): Number of results to return. Output - SearchEntitiesResponse - Response schema for the search_entities tool. ### get_entity Tool to get entities details from Artsdata Knowledge Graph by entity URI. Input parameters - uri (string, required): The Artsdata entity URI. Output - GetEntityResponse - Response schema for the get_entity tool. ### search_events Tool to search for events in the Artsdata knowledge graph by place, artist, organization, has_event_type_concept and language. Places, artists and organizations are Artsdata entity URIs; the has_event_type_concept filter takes concept URIs from the Artsdata Event Types vocabulary (adr:ArtsdataEventTypes), not schema.org classes. Resolve any name or label to its URI with search_entities first. Places must be venues; for events in a city or region, use sparql_query (see the places parameter). Input parameters - startDateFrom (string, optional): Optional lower bound filter. Includes events starting on or after this date/time. ISO 8601 format (YYYY-MM-DD or YYYY-MM-DDTHH:mm:ssZ). - startDateTo (string, optional): Optional upper bound filter. Includes events starting on or before this date/time. ISO 8601 format (YYYY-MM-DD or YYYY-MM-DDTHH:mm:ssZ). - places (array, optional): Optional list of Artsdata VENUE URIs (e.g., ["http://kg.artsdata.ca/resource/K11-2"]) for a theatre, hall, auditorium, room or stage. Only venue URIs work: a city, region or country (e.g. 'Toronto', 'Ontario', 'Canada') is not a venue and returns no events. First find the venue's URI with search_entities (types: ['http://schema.org/Place']) using the venue name. A venue URI also covers the rooms and halls inside it (schema:containedInPlace). To find events in a city or region, use sparql_query. Can be left empty. - artists (array, optional): Optional list of Artsdata artist URIs (e.g., ["http://kg.artsdata.ca/resource/K2-6574"]). If you only have a name, use search_entities (types: ['Person']) to find its URI first. Can be left empty. - organizations (array, optional): Optional list of Artsdata organization URIs (e.g., ["http://kg.artsdata.ca/resource/K5-72"]). If you only have a name, use search_entities (types: ['Organization']) to find its URI first. Can be left empty. - has_event_type_concept (array, optional): Optional list of concept URIs from the Artsdata Event Types vocabulary (http://kg.artsdata.ca/resource/ArtsdataEventTypes). Must be valid concept URIs (e.g., http://kg.artsdata.ca/resource/ClassicalMusicPerformance). Non-matching inputs—such as text labels, schema.org classes, or external vocabularies—will return zero results without throwing an error. Resolve any name or label to its URI with search_entities first - language (string, optional, default en): Language for the labels in the response (e.g. 'en', 'fr'); default is 'en'. This selects the output language only — it does NOT filter events by the language they are performed in. - limit (integer, optional, default 25, min 1, max 50): Number of results to return. Output - SearchEventsResponse - Response schema for the search_events tool. ### get_schema Returns the data model of the Artsdata Knowledge Graph: its entity classes (Event, Place, Organization, Person, LivePerformanceWork...), the properties of each with required/optional, cardinality, value types, linked classes and allowed values, the controlled vocabularies, the SPARQL prefixes, and conventions for querying. Derived from the Artsdata Ontology and the Artsdata CORE graph SHACL shapes. Call this before writing a SPARQL query to run with sparql_query against https://query.artsdata.ca/query, since Artsdata's model is not general knowledge. Takes no input. Input parameters - None Output - GetSchemaResponse - Response schema for the get_schema tool: the Artsdata data model (ontology + CORE graph SHACL shapes) compiled into a class-centric digest. IRIs are CURIEs over `prefixes` (or when no prefix fits); property paths are SPARQL 1.1 property paths. ### sparql_query Runs a read-only SPARQL 1.1 SELECT or ASK query against the Artsdata Knowledge Graph (https://query.artsdata.ca/query) and returns the result as SPARQL 1.1 Query Results JSON, exactly as the endpoint returns it: head.vars and results.bindings for SELECT, boolean for ASK. Call get_schema first and build the query from the classes, properties, prefixes and conventions it returns, since Artsdata's model is not general knowledge. Declare every PREFIX you use and always add a LIMIT: at most 1000 rows are returned (the rest is cut and truncated is true; page with OFFSET), and a query gets 25 seconds. If the endpoint rejects the query, its error message is returned so you can fix the query and retry. For the details of an entity URI in the results use get_entity; to find an entity's URI from its name use search_entities. Input parameters - query (string, required): A complete SPARQL 1.1 SELECT or ASK query, with a PREFIX declaration for every prefix it uses. Build it from the classes, properties and prefixes returned by get_schema, and add a LIMIT. Output - SparqlQueryResponse - Response schema for the sparql_query tool: the endpoint's SPARQL 1.1 Query Results JSON (https://www.w3.org/TR/sparql11-results-json/), as returned. A SELECT gives head.vars and results.bindings; an ASK gives boolean. ## MCP resources ### artsdata_dump - Title: Artsdata core minus provenance - URI: artsdata://dumps/core-minus-provenance/latest - MIME type: application/ld+json - Description: Manifest for the latest Artsdata core-minus-provenance dump (gzipped Turtle). Reading this resource returns JSON metadata (download URL, version, format) - not the dump itself. Download the file from `downloadUrl` to get the data. ## Notes for agents - Prefer MCP tools over scraping website HTML when possible. - Call get_schema before writing SPARQL for this graph model. - For raw graph queries, use sparql_query with explicit PREFIX declarations and LIMIT.